|
|
||||||||
Ann Thorac Surg 2002;73:892-898
© 2002 The Society of Thoracic Surgeons
a Service de Chirurgie Thoracique, Hôpital Européen Georges Pompidou, Paris, France
b Institut dAnatomie, UER Biomédicale des Saints Pères, Paris, France
Accepted for publication September 25, 2001.
* Address reprint requests to Dr Riquet, Service de Chirurgie Thoracique, Hôpital Européen Georges Pompidou, 20 rue Leblanc, 75015 Paris, France
e-mail: marc.riquet{at}hop.egp.ap-hop-paris.fr
| Abstract |
|---|
|
|
|---|
Methods. The study was performed on 530 adult cadavers. The lymphatics of different organs were catheterized and injected with a dye: lungs (n = 360), heart (n = 90), esophagus (n = 50), and diaphragm (n = 30). The lymphatic tributaries draining the lymph from these organs to the thoracic duct were dissected along their course to the thoracic duct and classified.
Results. The TD tributaries were observed in 147 cases: right lung (n = 46), left lung (n = 69), heart (n = 8), esophagus (n = 13), and diaphragm (n = 11). Connections with the TD were observed at its origin (n = 13), within the mediastinum (n = 87), and at the level of the TD arch (n = 47). Tributaries from the lung issued from lower paratracheal nodes 4 R (n = 14) and 4 L (n = 31), subaortic 5 (n = 4), subcarinal 7 (n = 18), pulmonary ligament 9 (n = 7), upper tracheal 2 L (n = 28), paraortic 6 (n = 11), and celiac nodes (n = 2). Tributaries from the heart connected with the TD in the mediastinum in 1 case (4 L) and with the TD arch in 7 cases. Tributaries from the esophagus connected with the thoracic duct within the mediastinum in 13 cases; anodal routes were frequent (n = 5). The TD tributaries from the diaphragm were observed in 11 cases, always connecting with the TD at its origin.
Conclusions. Injection of intrathoracic organs permits visualization of TD tributaries. These tributaries appear located at unchanging levels. Lymph of intrathoracic organs may thus drain into the general circulation through the TD. The tributaries may represent a potential route for tumor cells dissemination. When incompetent, due to valve insufficiency, they permit chylous lymph to backflow into the intrathoracic lymph nodes. Injury at this level may lead to intrathoracic chylous effusions.
| Introduction |
|---|
|
|
|---|
| Material and methods |
|---|
|
|
|---|
Lymphatic catheterization was performed with conventional equipment for pedal lymphography. The maneuver was simple to perform with magnification. For injection we used a dye consisting of a modified Gerotas medium with a blue or green colorant.
When studying the lungs, the subpleural lymphatics were catheterized directly. When possible we injected segments from either sides with a blue or green dye. After injection, the heart was removed through the pericardium and the mediastinum and the neck were dissected plane by plane, from the surface inward. When studying the heart, we opened the pericardium and injected the dye into the myocardium (green for the right ventricle and blue for the left ventricle). The lymph vessels were catheterized and reinjected. The dissection was then performed layer-by-layer within the neck and mediastinum. A similar approach was used when studying the diaphragm: catheterization was performed directly when the lymph vessels were visible, which was infrequent. When not visible, dye was injected into the tendinous part of the diaphragm, which by diffusion allowed visualization of small lymph vessels merging to form vessels amenable to direct catheterization. For the esophagus we proceeded differently. After removal of the sternocostal shield, the intrathoracic organs were eviscerated (passing behind the prevertebral ligament) from the skull base to the lumbar vertebrae. The esophagus of each subject was routinely injected alternatively with blue or green dyes. Injections were performed in the posterior aspect of the esophagus at different levels corresponding to the cervical esophagus and thoracic esophagus which was divided at three levels.
The quantity of injected dye varied from 1 to 5 mL, 2 mL on average. Injection was performed using a syringe and minimal pressure. Complete and perfect injections could be obtained with minimal doses of dyes, but frequently this quantity had to be adapted to each case because of technical problems (leakage). From the site of injection, the dye filled lymphatics and then the lymph nodes located within the mediastinum and the retroperitoneum. These lymph nodes are shown in Figure 1. The classification used was as Riquet and colleagues [6]. For better understanding it was associated to the regional lymph node classification for lung cancer staging reported in 1997 by Mountain and Dresler [7] when lymph nodes draining the lungs were encountered (Table 1, Fig 1). From the injected lymph nodes, the dye followed the lymphatic vessels upward (cephalad) and ultimately reached the jugulo-subclavian venous confluence. Some lymphatics (n = 147) connected with the thoracic duct along their ascending course to the neck.
|
|
| Results |
|---|
|
|
|---|
Tributaries from the lungs
The lymphatic vessels from the lung connected with the thoracic duct in 115 cases (Table 1, Fig 1). Connection occurred in the mediastinum in 73 cases, at the level of the arch of the TD in 40 cases, and at its origin in 2 cases. The connecting lymphatic vessels came from mediastinal lymph node chains in 112 cases and from the lung itself, without crossing any lymph node on their course in 3 cases. All of the latter 3 cases were located within the inferior pulmonary ligaments (Fig 1). From the right paratracheal nodes (Mountains 4 R [7]), the lymphatic tributaries to the thoracic duct traveled along the arch of the azygos vein, on the right of the trachea and esophagus. From the left superior bronchial nodes (Mountains 4 L [7]) the lymphatic vessels drained directly to the TD in the mediastinum (n = 34) (Fig 2)
or to the arch of the thoracic duct by the left recurrent chain (n = 28) (Fig 3).
From the aortic arch node (Mountains 5 [7]) the lymphatic vessels connected with the thoracic duct directly in the mediastinum traveling along the arch of the aorta (n = 4) (Fig 4)
or with the arch of the TD by the left anterior mediastinal lymph node chain (n = 11). From the nodes of the tracheal bifurcation (n = 18), the tributaries joined the TD either on the right or left of the esophagus (Fig 1).
|
|
|
|
|
|
|
| Comment |
|---|
|
|
|---|
A major objection of this technique may be that the networks so injected in cadavers cannot be postulated as drainage routes, but only as preferential pathways. However, these preferential pathways were found with great regularity from one subject to another in our study. The reason being that lymphatic vessels are valved [4, 8] and that these valves are still competent in the cadaver. Thus, in studies injecting the thoracic duct itself, it was always necessary to perform injections caudally [3, 5] and the dye never flowed back into the tributaries because of valve competency. The flow thus canalized after catheterization and direct injection is therefore comparable to that occurring in the living human subject.
Direct injection has the further advantage of being faithful to the injected lymphatic system [6], whatever the amount of injected dye. When the venous circulation is reached, the network that is revealed by the injection will not be modified further. In completely injected cases with venous connection, the quantity of dye was often small, whereas in other cases, it was necessary to inject more dye because of dye loss within the interstitium. Dye loss was due to its diffusion through the walls of the lymphatics or leakage within the mediastinum, impossible to detect during injection. Thus, it was not possible to obtain a complete injection in all cases. This explains our success rate of 27.7% (147 of 530), which increases to 32.3% for the lung and diaphragm (121 of 390) and as low as 15% for heart and esophagus (21 of 140).
It is important to stress that the number of lymph vessels connecting with the TD within the mediastinum was probably underevaluated. This was mainly due to the above mentioned difficulties encountered in numerous cases, which did not permit complete injection of the lymphatic system. Another reason for underevaluation was the fact that the injections performed within the same organ were few (most often only one) and did not reflect the richness of its lymphatic drainage. It would not seem an exaggeration to postulate that the TD tributaries observed are present and constant in all individuals.
Anatomic pattern of the tributaries
The distribution of the TD tributaries was easy to describe; however, it is important to stress three points.
First, the TD tributaries drain lymph from lymph node groups known as the first lymph node group on the course of lymph drainage from intrathoracic organs: the celiac and paraaortic nodes receive lymph from the lower lobes of the lungs and the diaphragm; the inferior pulmonary ligament nodes (Mountains 9 [7]) from the lower lobes of the lung; the esophageal nodes (Mountains 8 [7]), from the esophagus; the aortic arch nodes (Mountains 5 [7]), paraortic (ascending aorto or phrenic) (Mountains 6 [7]), and the supra bronchial nodes (Mountains 4 L [7]) draining the lymph from both lungs, and so on. This is clearly demonstrated in our study.
Second, the same lymph node groups may receive lymph from other organs, thus being the second lymph node group on the course of the lymphatic drainage. It is likely that the TD tributaries issuing from these lymph node groups may also drain the lymph of these organs: TD tributaries from the nodes of the tracheal bifurcation (Mountains 7 [7]) draining both lungs, but also the lymph from the esophagus and diaphragm [9, 10]; TD tributaries from the lower part of the right paratracheal nodes chains (Mountains 4 [7]) draining the lymph from both lungs, but also the lymph of the diaphragm, esophagus, and most of the lymph of the left ventricle of the heart [11]; TD tributaries from the left superior bronchial nodes (Mountains 4 L [7]), draining the lymph from both lungs, but also from the esophagus, the heart, and the diaphragm through the right paratracheal nodes chains [911].
Finally, some TD tributaries may directly drain lymph from an organ without crossing any lymph node (anodal route) along its course. This was observed for the diaphragm, the esophagus, and the lower lobes of the lungs. This point may play a particular role in prognosis of cancers involving these organs.
TD lymph tributaries and their possible role in thoracic oncology
Each of the anatomic lymph node chains observed and some groups of lymph nodes of the "regional lymph node classification for lung cancer staging" may empty lymph into the general circulation either directly into the cervical venous junction or through the thoracic duct tributaries. These chains also interanastomose with each other. The poor prognosis observed in non-small cell lung cancer when two ipsilateral mediastinal lymph node stations or chains are tumoral (N2) [12] may be explained by these anastomoses. The chain first injected has already abundantly poured out cancerous cells into the blood circulation when a second chain is metastatic. Because of this double mode of drainage of lymph into the bloodstream, each anatomic lymph node chain is thus working as a "functional entity." Moreover, death in these N2 cases appears primarily linked to systemic metastases, the incidence of which exceeds 90% [13]. This also supports the hypothesis that the mediastinal lymph node chain works as a functional entity and thus as the ultimate barrier before distant spread.
In addition, tributaries may connect with the TD without crossing any lymph node during their course (anodal route). This represents a direct mode of spread of tumor cells into the general circulation. This may help explain the poor prognosis of esophageal cancer and the possibility of distant metastases in non-small cell lung cancer, even when there is no lymph node involvement. It may also help understand the poor prognosis of peripheral non-small cell lung cancer invading the lung visceral pleura. Brewer [14] explained that such visceral pleura invasion produces dissemination of cancer cells throughout the pleural cavity in the stream of the pleural fluid. Preformed stomas that connect subpleural lymphatics with the pleural space could further account for lymphatic and systemic tumor cell dissemination [15]. The diaphragmatic tributaries that directly empty into the TD origin, as we observed, may provide an explanation for this potential mode of spread.
TD lymph tributaries and intrathoracic chylous effusions
Postoperative chylothorax is extremely rare after pulmonary resections even when a wide mediastinal lymph node dissection is performed. This is due to valve competency of the lymphatic vessels. Chylothorax is possible only in cases of valve insufficiency. In such cases, the site of the leak when identified is generally found at the level of the lymph nodes giving issue to the TD tributaries we describe. Such examples may retrospectively be found in the literature: right paratracheal (Mountains 4 R [7]) by 3 p [16]; inferior pulmonary ligament (Mountains 9 [7])B [16, 17]; paraesophageal (Mountains 8 [7]) [16]; bifurcation nodes (Mountains 7 [7]) [16]; superior left bronchial (Mountains 4 L [7]) [16, 17]. The case observed by Van Mulders and associates [18], where "the leak could be seen oozing out posteriorly adjacent to the aorta at the origin of the intercostal arteries where a pleural flap had been removed to cover the bronchial stump" was probably located at the level of lymph vessels represented on Figure 4 and issuing from the "aortic nodes" (Mountains 5 [7]).
Lymphangiography is the only investigation permitting the demonstration of backflow from the TD and thus the leak. However, it may fail to identify the leak as reported by Vallieres and colleagues [19]. In this last case, contrast medium was only seen within a network of small lymphatics and nodes in the posterior and midmediastinum, extending to the left supraclavicular fossa; nevertheless, this was the indication of backflow. In a patient reported by Akaogi and coworkers [20], the lymphangiogram revealed that chylous leakage (perfectly visible on the lymphangiogram) occurred at the level of the azygos vein (right paratracheal nodes = Mountains 4 R [7]).
In cases of left pulmonary resection, chylous leak may occur at the level of the anterior mediastinal chain [16], and at the origin of the left subclavian artery [21]. In such cases backflow occurs from the TD arch. At this same level occur most chylous leaks observed after coronary artery bypass grafting and left internal thoracic artery harvest. This clearly appears in the reports of Chaiyaroj [22] and Di Lello [23] and their colleagues describing a tributary at this level. It is not as evident in the reports of Bogers [24] and Inderbitzi [25] and associates, where although the leak was located at the same level, it was not as well identified.
Chylothorax observed after median sternotomy for cardiac procedures may be due to injuries to the upper part of the right efferent lymphatic trunk, a lymph tributary to the arch of the thoracic duct. Joyce and colleagues [26] observed at reexploration a continuous flow of chylous lymph coming from a lymphatic channel lying on the left side in the area of the thymus in 2 patients.
The injury of an incompetent right efferent lymphatic trunk along its course within the pericardium between the ascending aorta and the pulmonary trunk may also account for postoperative chylopericardium. Although this is difficult to prove from rare cases of postoperative chylopericardium reported in the literature, a case published by Pollard and associates [27] merits mention: at operation, the surgeons impression was that the number of superior mediastinal lymphatics in the region of the ascending aorta and main pulmonary artery was increased.
Other types of intrathoracic operations may lead to chylothorax. Left chylothorax may be observed after closure of a patent ductus arteriosus, probably due to injury of aortic and left suprabronchial tributaries (5 and 4 L). Right chylothorax may be observed after Blalok- Taussig and Waterston operations, probably due to injury of the azygos vein tributaries (4 R).
On the contrary, chylothorax after esophagectomy may be mainly related to direct injury of the thoracic duct due to the vicinity of both structures [28]. The thoracic duct may also be directly at risk during operation of the aortic arch.
In conclusion, injection of intrathoracic organs permits visualization of lymph tributaries to the thoracic duct. These tributaries are located at constant levels. The lymph flows from the organs into the general circulation through the TD. In so doing, they are a potential mode of tumor cell dissemination. When incompetent, due to valve insufficiency, they permit the chylous lymph to backflow into the intrathoracic lymph nodes and into the organs; injury at their level may lead to intrathoracic chylous effusions.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
D. Stewart, T. Chico, and R. Ackroyd Chylopericardial tamponade complicating oesophago-gastrectomy Interactive CardioVascular and Thoracic Surgery, January 1, 2009; 8(1): 176 - 177. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. K. Choong, C. Martinez, and H. B. Barner Chylothorax After Internal Thoracic Artery Harvest Ann. Thorac. Surg., April 1, 2006; 81(4): 1507 - 1509. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Falode, I. Hunt, and C. P Young Chylothorax after coronary artery bypass surgery J R Soc Med, July 1, 2005; 98(7): 314 - 315. [Full Text] [PDF] |
||||
![]() |
M. Riquet, J. Assouad, N. D'Attellis, and I. Gandjbakhch Chylothorax and re-expansion pulmonary edema following myocardial re-vascularization: role of lymph vessel insufficiency Interactive CardioVascular and Thoracic Surgery, September 1, 2004; 3(3): 423 - 425. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Murakami and I. Taniguchi Histologic Heterogeneity and Intranodal Shunt Flow in Lymph Nodes from Elderly Subjects: A Cadaveric Study Ann. Surg. Oncol., March 1, 2004; 11(3_suppl): 279S - 284S. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Sharma, P. Fidias, L. A. Hayman, S. L. Loomis, K. H. Taber, and S. L. Aquino Patterns of Lymphadenopathy in Thoracic Malignancies RadioGraphics, March 1, 2004; 24(2): 419 - 434. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Le Pimpec-Barthes, N. D'Attellis, A. Dujon, P. Legman, and M. Riquet Chylothorax complicating pulmonary resection Ann. Thorac. Surg., June 1, 2002; 73(6): 1714 - 1719. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |